Daily Papers of 2025-04-16

  1. xVerify: Efficient Answer Verifier for Reasoning Model Evaluations 83 upvotes, #1 of 2025-04-16
  2. Genius: A Generalizable and Purely Unsupervised Self-Training Framework For Advanced Reasoning 53 upvotes, #2 of 2025-04-16
  3. Seedream 3.0 Technical Report 45 upvotes, #3 of 2025-04-16
  4. How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients 39 upvotes, #4 of 2025-04-16
  5. Heimdall: test-time scaling on the generative verification 32 upvotes, #5 of 2025-04-16
  6. Pixel-SAIL: Single Transformer For Pixel-Grounded Understanding 28 upvotes, #6 of 2025-04-16
  7. TextArena 27 upvotes, #7 of 2025-04-16
  8. Efficient Reasoning Models: A Survey 18 upvotes, #8 of 2025-04-16
  9. NormalCrafter: Learning Temporally Consistent Normals from Video Diffusion Priors 17 upvotes, #9 of 2025-04-16
  10. The Scalability of Simplicity: Empirical Analysis of Vision-Language Learning with a Single Transformer 15 upvotes, #10 of 2025-04-16
  11. DataDecide: How to Predict Best Pretraining Data with Small Experiments 15 upvotes, #10 of 2025-04-16
  12. ReZero: Enhancing LLM search ability by trying one-more-time 14 upvotes, #12 of 2025-04-16
  13. A Minimalist Approach to LLM Reasoning: from Rejection Sampling to Reinforce 14 upvotes, #12 of 2025-04-16
  14. Efficient Process Reward Model Training via Active Learning 13 upvotes, #14 of 2025-04-16
  15. Efficient Generative Model Training via Embedded Representation Warmup 12 upvotes, #15 of 2025-04-16
  16. SimpleAR: Pushing the Frontier of Autoregressive Visual Generation through Pretraining, SFT, and RL 12 upvotes, #15 of 2025-04-16
  17. D^2iT: Dynamic Diffusion Transformer for Accurate Image Generation 11 upvotes, #17 of 2025-04-16
  18. RealHarm: A Collection of Real-World Language Model Application Failures 11 upvotes, #17 of 2025-04-16
  19. DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning 11 upvotes, #17 of 2025-04-16
  20. VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge 10 upvotes, #20 of 2025-04-16
  21. Efficient Hybrid Language Model Compression through Group-Aware SSM Pruning 10 upvotes, #20 of 2025-04-16
  22. AI-University: An LLM-based platform for instructional alignment to scientific classrooms 8 upvotes, #22 of 2025-04-16
  23. PVUW 2025 Challenge Report: Advances in Pixel-level Understanding of Complex Videos in the Wild 6 upvotes, #23 of 2025-04-16
  24. Diffusion Distillation With Direct Preference Optimization For Efficient 3D LiDAR Scene Completion 5 upvotes, #24 of 2025-04-16
  25. Multimodal Long Video Modeling Based on Temporal Dynamic Context 4 upvotes, #25 of 2025-04-16
  26. Adaptive Computation Pruning for the Forgetting Transformer 3 upvotes, #26 of 2025-04-16
  27. Summarization of Multimodal Presentations with Vision-Language Models: Study of the Effect of Modalities and Structure 3 upvotes, #26 of 2025-04-16
  28. LazyReview A Dataset for Uncovering Lazy Thinking in NLP Peer Reviews 3 upvotes, #26 of 2025-04-16
  29. Aligning Generative Denoising with Discriminative Objectives Unleashes Diffusion for Visual Perception 2 upvotes, #29 of 2025-04-16
  30. Change State Space Models for Remote Sensing Change Detection 1 upvotes, #30 of 2025-04-16

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.